Archives of Cardiovascular Disease (2012) 105, 153—159

Available online at

www.sciencedirect.com

REVIEW

Transcatheter aortic valve implantation:The evolution of prostheses, delivery systems andapproaches

Implantation de valves aortiques par voie percutanée : évolutionstechnologiques

John G. Webb ∗, Ronald K. Binder

McLeod professor and British Columbia director structural heart intervention, University ofBritish Columbia, St. Paul’s Hospital, 1081, Burrard Street, Vancouver, BC, V6Z 1Y6, Canada

Received 20 January 2012; received in revised form 1st February 2012; accepted 2 February 2012Available online 16 March 2012

KEYWORDSTranscatheter aorticvalve implantation(TAVI);Aortic stenosis;Medtronic CoreValve;SAPIEN valves

Summary It is two decades since the first report of transcatheter implantation of a stentedaortic valve in an animal. The first implantation of a transcatheter aortic valve in a human wasaccomplished just one decade ago dramatically demonstrating the promise and feasibility ofthis new therapy. Over the past 10 years, there have been rapid developments in valves, deli-very systems and technical approaches. Today, transcatheter valve implantation is a technicalpossibility for the great majority of patients with aortic stenosis. The next 10 years may wellsee this become the dominant therapy for aortic stenosis.© 2012 Elsevier Masson SAS.

MOTS CLÉSValve aortique ;Transcatheter ;

Résumé Deux décennies se sont écoulées depuis la première implantation d’une valve sten-tée chez l’animal. La première implantation d’une valve aortique percutanée chez l’homme aété réalisée, il y a juste dix ans, avec une faisabilité démontrée et des perspectives promet-teuses. Ces dix dernières années ont connu un développement technologique rapide que ce

Open access under CC BY-NC-ND license.

Rétrécissement

aortique ;Remplacementvalvulaire aortique

soit sur les valves elles-mêmes,

l’implantation de valves aortiqude patients porteurs d’un rétrécannées, cette technique devienn© 2012 Elsevier Masson SAS.

Abbreviations: TAVI, Transcatheter aortic valve implantation; THV, Tr∗ Corresponding author.

E-mail address: john.webb@vch.ca (J.G. Webb).

1875-2136 © 2012 Elsevier Masson SAS.

doi:10.1016/j.acvd.2012.02.001Open access under CC BY-NC-ND licens

Cet a

les dispositifs de délivrance ou les voies d’abord. Aujourd’hui,es percutanées est techniquement accessible à une majoritéissement aortique. Il est possible que dans les dix prochainese le traitement prédominant du rétrécissement aortique.

anscatheter heart valve.

e.

rticle est publié en Open Access sous licence CC BY-NC-ND.

154

Figure 1. Early catheter-mounted valves. Parachute-like valvesoffered little resistance to the flow of blood in one direction, butobstructed the flow in the other. Animal studies suggested benefitin the setting of aortic regurgitation.Re

B

AcdorOeaiw

tdsioss[wTibovPsppi

at

ad

E

B

TIscrqvpl

ltamlavFmgeaa

C

IeGntbccwtsascaI2

[imbuprocess is not without complications [15]. Coronary obstruc-

eprinted with permission from Davies. 1965 [1] (A) and Phillipst al., 1976 [4] (B).

ackground

lmost 50 years ago, in 1965, Hywel Davies placed aatheter-mounted valve via the femoral artery into theescending aorta of a dog [1]. The parachute-like valve wasriented so as to passively open in diastole and obstructetrograde flow in the patients with aortic regurgitation.ver the subsequent few decades, a number of groupsxperimented with various catheter-mounted passive andctive temporary valves (Fig. 1). These valve designs werentended as temporary palliation for aortic regurgitation andere never utilized in humans [2—4].

The first catheter intervention specifically targeted athe stenotic aortic valve was aortic balloon valvuloplasty;eveloped initially as a therapy for congenital aortictenosis, and described for degenerative aortic stenosisn adults in 1986 by Alain Cribier [5]. The feasibilityf dilation of the stenotic calcific valve provided theeed for future developments. In 1992, Henning Ander-en described the first transcatheter aortic stent valve6]. This was constructed of a handmade wire frameithin which was sewn a porcine aortic valve (Fig. 2).he assembly was crimped onto a balloon catheter and

mplanted transarterially into a pig. Subsequently, a num-er of groups, including our own, pursued the developmentf a practical and reliable transcatheter stent-mountedalve suitable for implantation in humans [7]. In 2000,hilippe Bonhoeffer described a bovine jugular venous valveewn within a large-diameter stent, which was implantedercutaneously in children with pulmonary conduits [8];roving the feasibility of transcatheter valve implantationn humans.

However, it was in 2002, 10 years ago, that Alain Cribierccomplished the first transcatheter aortic valve implan-ation (TAVI) for aortic stenosis [9]; opening the door to

tvt

J.G. Webb, R.K. Binder

new age in the management of this relatively commonisease.

volution of transcatheter valves

alloon-expandable valves

he original Cribier-Edwards valve (Edwards Lifesciencesnc., CA, USA) was constructed from a laser-cut stainlessteel tubular frame within which was sewn valve leafletsonstructed from equine pericardium, with the inflow cove-ed with a fabric sealing cuff (Fig. 3A). This valve was subse-uently modified as the Edwards SAPIEN transcatheter heartalve (THV) (Fig. 3B), incorporating more durable bovineericardium, and a higher sealing cuff to reduce paravalvu-ar leaks [10].

The SAPIEN valve was subsequently superseded by theow-profile Edwards SAPIEN-XT valve (Fig. 3C); in whichhe stainless steel was replaced with a cobalt chromiumlloy frame [10,11]. This alloy allows for a thinner, stronger,ore open and compressible frame, while the reengineered

eaflets ensure valve closure even when closing pressuresre low, along with increased durability. Importantly, thisalve was designed to allow the use of smaller diameter 18rench delivery systems. The SAPIEN-XT valve is currentlyanufactured with diameters of 20, 23, 26 and 29 mm. Next-

eneration SAPIEN valve systems are currently undergoingvaluation, and will incorporate additional leaflet, framend sealing enhancements to further reduce delivery profilend paravalvular leaks.

oreValve

n 2005, following initial implants in India, early Germanxperience with the CoreValve (Fig. 4) was reported byrube et al. [12]. The valve frame was constructed of niti-ol; a nickel-titanium alloy that can be manufactured so aso be malleable when cool, but to become relatively rigid atody temperature. The valve leaflets and annular seal wereonstructed of porcine pericardium. The valve is cooled andompressed within a delivery catheter, which is then placedithin the diseased valve. As a covering sheath is withdrawn,

he valve is released and expands, returning to its presethape. A long multi-staged frame is anchored within theortic annulus as with the SAPIEN valves, but also extendsuperiorly to anchor in the supracoronary aorta. The supra-oronary portion serves to self-centre the valve and providesn additional point of fixation. The CoreValve (Medtronicnc., MN, USA) is currently available with diameters of 26,9 and 31 mm.

Only limited non-randomized comparisons are available13,14]. Deployment of the CoreValve device may be morentuitive, and does not require rapid pacing, while deploy-ent of the SAPIEN device may be more targeted and clinicalenefit is better documented. The CoreValve device can,p to a point, be repositioned or retrieved, although this

ion is rare, but more of a concern with the SAPIEN-typealves. Atrioventricular block requiring pacemaker implan-ation is much more common with CoreValve. Both valves

Evolution of TAVI 155

Figure 2. The Andersen valve. The valve was constructed of a handmade wire frame to which was sewn a porcine valve.Reprinted with permission from Dr. Henning Andersen.

PIEN (B) and SAPIEN-XT (C) valves.

Figure 3. Balloon-expandable valves. The Cribier-Edwards (A), SAReprinted with permission.

can offer dramatic clinical benefit and both systems con-tinue to improve.

Newer valves

Several newer transcatheter valves (Fig. 5) are in variousstages of evaluation. Some are in clinical use, although expe-rience remains very limited [16—18]. Generally, these newervalves attempt to improve on the widely available SAPIEN-XTand CoreValve devices by enhancing deliverability, position-ing, sealing, or facilitating repositioning or removal.

Some valves have unique expansion mechanisms. TheLotusTM valve (Boston Scientific Inc., MN, USA) (Fig. 5A)frame is constructed of woven nitinol wires which, whentensioned, cause the valve to decrease in height whileincreasing in diameter and rigidity. The Direct FlowTM valve

(Direct Flow Medical Inc., CA, USA) (Fig. 5B) has a tubularfabric frame that is inflated with a rapid-setting polymer-izing agent. The CENTERA valve (Edwards Lifesciences Inc.,CA, USA) (Fig. 5C) is released from its constrained state by

Figure 4. Medtronic CoreValve.

156 J.G. Webb, R.K. Binder

Figure 5. Newer valves: LotusTM valve (Boston Scientific Inc., MN, USA) (A); Direct FlowTM valve (Direct Flow Medical Inc., CA, USA) (B);CENTERATM valve (Edwards Lifesciences Inc., CA, USA) (C); PorticoTM valve (St. Jude Medical Inc., MN, USA) (D); EngagerTM valve (MedtronicInc., MN, USA) (E); JenaClipTM valve (JenaValve Inc., Munich, Germany) (F); AcurateTM valve (Symetis, Ecublens, VD, Switzerland) (G); Brailev

aut

IVipS((meamatvac

sfiCnBs

E

T

TnltaCblpt

F

ToialTs

alve (Braile Biomédica, Brazil) (H).

n electric motor. However, for the most part, these valvestilize self-expanding heat-set nitinol frames that facilitatehinner delivery systems and repositioning.

Some valves, such as the PorticoTM (St. Jude Medicalnc., MN, USA) (Fig. 5D) and AcurateTM (Symetis, Ecublens,D, Switzerland) (Fig. 5G) valves, are long and extend

nto the supracoronary aorta to assist with alignment androvide additional fixation, as does the CoreValve device.ome, such as the EngagerTM (Medtronic Inc., MN, USA)Fig. 5E), Acurate (Symetis, Ecublens, VD, Switzerland)Fig. 5G), and JenaClipTM (JenaValve Inc., Munich, Ger-any) (Fig. 5F) valves, incorporate flexible arms that

xtend above the native leaflets to facilitate positioningnd rotational orientation in relation to the native com-issures and coronaries. These newer valves are generally

iming to match the 18 French diameter of current sys-ems, although 16 French and even 14 French systems areery near. The Braile valve (Braile Biomédica, Brazil) is

balloon-expandable valve, perhaps representing a lowerost alternative.

We can expect two new systems from Edwards Life-ciences Inc. to undergo CE mark evaluation in the nearuture. The balloon-expandable SAPIEN 3TM valve willncorporate features to improve paravalvular sealing. The

TM

ENTERA valve (Fig. 5C) will offer a self-expandable alter-ative with an electric motor assisted release mechanism.oth will be compatible with 14 to 16 French expandableheath delivery systems.

asd

volution of approaches

ransvenous

he first TAVI procedures were performed with percuta-eous femoral venous access, with transseptal access to theeft atrium, mitral valve, left ventricle, and then antegradehrough the aortic valve [9,19]. The femoral vein was easilyble to accommodate the large sheath required for the bulkyribier-Edwards valve crimped onto a standard valvuloplastyalloon. However, the procedure was complex, difficult toearn, and complications were common. Consequently thisrocedure was largely abandoned with the advent of theransarterial procedure in 2005.

emoral arterial

he retrograde transarterial femoral procedure was devel-ped concurrently by our group in Canada and Grube et al.n Germany [20,21]. Transfemoral arterial access resulted in

more reproducible and generalizable procedure, and wasargely responsible for the rapid uptake of TAVI worldwide.he main initial limitation was the size of early generationystems, ranging from 22 to 25 French in diameter.

The large-diameter CoreValve delivery system was rel-tively well suited to a reduction in profile, and 18 Frenchystems soon became available. Significant reductions in theiameter of the Edwards systems took longer due to the

Evolution of TAVI 157

Figure 7. The Edwards delivery systems. Retroflex deliverycSp

aC(rt

rrrdschit

A

Tmvteatpa[

Hwspv

pwell to this approach [25]. More recently, several new self-

Figure 6. Medtronic CoreValve and the Accutrak delivery system.

additional bulk of the deployment balloon. However, thisproblem was largely solved with the advent of the currentlow-profile NovaflexTM system, with which the valved stentwas introduced into the sheath first and only later mountedon the balloon after both had been passed into the aorta.

Both delivery systems have undergone a number of ite-rative changes over the past 7 years. In addition to areduction in profile, the CoreValve delivery system hasincorporated the proprietary Accutrak stability layerTM tech-nology (Fig. 6), which markedly reduced the forces requiredto retract the covering sheath, facilitating relatively moreaccurate placement of the valve.

The Edwards system evolved from a simple valvuloplastyballoon with a deflectable pusher RetroFlexTM catheter(Fig. 7A), to the RetroFlex 2 catheter that incorporateda tapered nosecone, then to the more refined RetroFlex3 catheter (Fig. 7B). The nosecone greatly increased theease of passage through the aortic arch and stenotic nativevalve. The Novaflex (Fig. 7C) and the, even newer and yetmore refined, Novaflex plus (Fig. 7D) catheters incorporatemechanisms for aligning the SAPIEN-XT valve on the balloon,allowing insertion through smaller sheaths. Numerous otherminor modifications allow more controlled and reliable pas-sage through the aorta, positioning and valve deployment.

Both the Edwards and the Medtronic systems initiallyused large 22 to 25 French sheath introducers designed forthe introduction of aortic endografts. The CoreValve sys-tem is currently reliant on off-the-shelf 18 French sheaths(Check-FloTM, Cook Medical Inc., IN, USA; UltimumTM, St.Jude Medical Inc., MN, USA; and DrySheathTM, Gore Medi-cal Inc., AZ, USA). Edwards Lifesciences developed its ownline of sheath introducers and dilators specifically designedfor this purpose. Recently, a 14 French expandable sheathhas become available (SoloPathTM, Onset Medical Corp., CA,

USA). Once introduced into the artery, a balloon bondedto the obturator is used to dilate the stent-like frameof the sheath to over 18 French, minimizing the risk of

eaJ

atheter with the Cribier-Edwards valve (A); Retroflex 3 with theAPIEN valve (B); Novaflex with the SAPIEN-XT valve (C); Novaflexlus (D).

rterial injury at the time of sheath insertion. BothoreValve and SAPIEN versions are available. The EsheathTM

Edwards Lifesciences Inc., CA, USA) incorporates a fold thatuns the length of the 14, 16 or 18 French sheath and dilatesransiently as the SAPIEN-XT valve is passed through it.

With reduced sheath diameters and increased expe-ience, open cutdown is progressively giving way tooutine fully percutaneous access. Most operators cur-ently use preclosure with either the ProstarTM or ProglideTM

evices (Abbott Medical Inc., IL, USA). With lower-profileystems, improved screening and increased expertise, vas-ular complications are falling markedly in experienced,igh volume centres [22]. Femoral transarterial accesss currently the preferred default strategy in most cen-res.

pical

he first transapical valve implantation in an animalodel was performed utilizing an early self-expanding

alve in Vancouver in 2000 [7]. Walther et al. implantedhe Cribier-Edwards valve with the left ventricular apexxposed through a median sternotomy in Leipzig in 2006,lthough paravalvular regurgitation required conversiono conventional surgery in a case [23]. Successful off-ump implants using intercostal access were successfullyccomplished by the Vancouver group shortly afterwards24].

Initially, the transfemoral Retroflex system was utilized.owever, the purpose-specific Ascendra delivery catheteras developed shortly afterwards. This has now been super-

eded by the Ascendra 2 system (Fig. 8), which is lowerrofile, easier to de-air, can accommodate the SAPIEN-XTalve, and facilitates use by a single operator.

Although a small number of CoreValve implants have beenerformed transapically, this system does not lend itself

xpandable systems have been designed to take specificdvantage of an apical approach; implanting the Engager,enaValve, Acurate and Portico valves.

158

F

T

Avwnawvs

poniata

T

DpocorpEahiba

C

Ifiams

D

J

t

R

[

[

[

[

[

[

Interv 2010;3:531—6.

igure 8. Ascendra 2 transapical valve delivery catheter.

ransaxillary/subclavian

ccess to the axillary artery and passage through the subcla-ian artery often offers an appealing alternative in patientsith iliofemoral disease [26,27]. The short distance andon-tortuous course to the aortic valve proved particularlydvantageous for the CoreValve procedure. Initial concernsith respect to the Novaflex system, due to the need foralve alignment in a relatively straight segment of the aorta,eem to have been allayed.

Access to the axillary artery has generally been accom-lished in an open fashion, due to the thin friable wallf this artery, although the feasibility of reliable percuta-eous access and closure has been demonstrated. However,t is necessary to assure a fallback should initial attemptst haemostasis fail in this non-compressible artery, makinghis somewhat more complex than is the case with femoralccess.

ransaortic

irect access to the ascending aorta through a smallarasternal incision is much more within the comfort zonef the cardiac surgeon uninitiated to left ventricular api-al access [28,29]. Although this avoids the risk of apicalr peripheral vascular injury, there are still concerns withespect to aortic injury or atheroembolism. Although therocedure has generally been performed with the standarddwards or Medtronic transarterial delivery system, we cannticipate purpose-specific systems that are shorter andave lower profiles to facilitate transaortic retrograde cross-ng of the aortic valve. It seems likely that this approach willecome increasingly popular in patients with compromisedrterial access.

onclusions

n the decade since transcatheter valve implantation wasrst accomplished in man, there have been tremendous

dvances. It is possible to speculate that the next 10 yearsay well see TAVI become the dominant therapy for aortic

tenosis.

[

J.G. Webb, R.K. Binder

isclosure of interest

ohn G. Webb: Consultant to Edwards Lifesciences.Ronald K. Binder: Recipient of an unrestricted grant from

he Swiss National Foundation.

eferences

[1] Davies H. Catheter-mounted valve for temporary relief of aor-tic insufficiency. Lancet 1965;285:250.

[2] Moulopoulos SD, Anthopoulos L, Stamatelopoulos S, Ste-fadouros M. Catheter-mounted aortic valves. Ann Thorac Surg1971;11:423—30.

[3] Matsubara T, Yamazoe M, Tamura Y, et al. Balloon catheter withcheck valves for experimental relief of acute aortic regurgita-tion. Am Heart J 1992;124:1002—8.

[4] Phillips SJ, Ciborski M, Freed PS, Cascade PN, Jaron D. Atemporary catheter-tip aortic valve: hemodynamic effectson experimental acute aortic insufficiency. Ann Thorac Surg1976;21:134—7.

[5] Cribier A, Savin T, Saoudi N, Rocha P, Berland J, Letac B. Percu-taneous transluminal valvuloplasty of acquired aortic stenosisin elderly patients: an alternative to valve replacement? Lancet1986;1:63—7.

[6] Andersen HR, Knudsen LL, Hasenkam JM. Transluminal implan-tation of artificial heart valves. Description of a newexpandable aortic valve and initial results with implanta-tion by catheter technique in closed chest pigs. Eur Heart J1992;13:704—8.

[7] Webb JG, Munt B, Makkar RR, Naqvi TZ, Dang N. Percutaneousstent-mounted valve for treatment of aortic or pulmonaryvalve disease. Catheter Cardiovasc Interv 2004;63:89—93.

[8] Bonhoeffer P, Boudjemline Y, Saliba Z, et al. Percutaneousreplacement of pulmonary valve in a right-ventricle topulmonary-artery prosthetic conduit with valve dysfunction.Lancet 2000;356:1403—5.

[9] Cribier A, Eltchaninoff H, Bash A, et al. Percutaneous tran-scatheter implantation of an aortic valve prosthesis for calcificaortic stenosis: first human case description. Circulation2002;106:3006—8.

10] Webb JG, Altwegg L, Masson JB, Al Bugami S, Al Ali A, BooneRA. A new transcatheter aortic valve and percutaneous valvedelivery system. J Am Coll Cardiol 2009;53:1855—8.

11] Nietlispach F, Wijesinghe N, Wood D, Carere RG, Webb JG. Cur-rent balloon-expandable transcatheter heart valve and deliv-ery systems. Catheter Cardiovasc Interv 2010;75:295—300.

12] Grube E, Laborde JC, Zickmann B, et al. First report on a humanpercutaneous transluminal implantation of a self-expandingvalve prosthesis for interventional treatment of aortic valvestenosis. Catheter Cardiovasc Interv 2005;66:465—9.

13] Moat NE, Ludman P, de Belder MA, et al. Long-term out-comes after transcatheter aortic valve implantation in high-riskpatients with severe aortic stenosis: the UK TAVI (United King-dom Transcatheter Aortic Valve Implantation) Registry. J AmColl Cardiol 2011;58:2130—8.

14] Eltchaninoff H, Prat A, Gilard M, et al. Transcatheter aor-tic valve implantation: early results of the FRANCE (FRenchAortic National CoreValve and Edwards) registry. Eur Heart J2011;32:191—7.

15] Geisbusch S, Bleiziffer S, Mazzitelli D, Ruge H, Bauernschmitt R,Lange R. Incidence and management of CoreValve dislocationduring transcatheter aortic valve implantation. Circ Cardiovasc

16] Falk V, Schwammenthal EE, Kempfert J, et al. New anatom-ically oriented transapical aortic valve implantation. AnnThorac Surg 2009;87:925—6.

[

[

[

[

[

[

Evolution of TAVI

[17] Treede H, Tubler T, Reichenspurner H, et al. Six-monthresults of a repositionable and retrievable pericardial valvefor transcatheter aortic valve replacement: the Direct FlowMedical aortic valve. J Thorac Cardiovasc Surg 2010;140:897—903.

[18] Schofer J, Schluter M, Treede H, et al. Retrograde transar-terial implantation of a nonmetallic aortic valve prosthesisin high-surgical-risk patients with severe aortic stenosis: afirst-in-man feasibility and safety study. Circ Cardiovasc Interv2008;1:126—33.

[19] Cribier A, Eltchaninoff H, Tron C, et al. Early experiencewith percutaneous transcatheter implantation of heart valveprosthesis for the treatment of end-stage inoperable patientswith calcific aortic stenosis. J Am Coll Cardiol 2004;43:698—703.

[20] Webb JG, Chandavimol M, Thompson CR, et al. Percutaneousaortic valve implantation retrograde from the femoral artery.Circulation 2006;113:842—50.

[21] Grube E, Laborde JC, Gerckens U, et al. Percutaneous implan-tation of the CoreValve self-expanding valve prosthesis inhigh-risk patients with aortic valve disease: the Siegburg first-

in-man study. Circulation 2006;114:1616—24.

[22] Toggweiler S, Gurvitch R, Leipsic J, et al. Percutaneous aorticvalve replacement vascular outcomes with a fully percutaneousprocedure. J Am Coll Cardiol 2012;59:113—8.

[

159

23] Walther T, Falk V, Borger MA, et al. Minimally invasive transapi-cal beating heart aortic valve implantation — proof of concept.Eur J Cardiothorac Surg 2007;31:9—15.

24] Ye J, Cheung A, Lichtenstein SV, et al. Transapical aor-tic valve implantation in humans. J Thorac Cardiovasc Surg2006;131:1194—6.

25] Lange R, Schreiber C, Gotz W, et al. First successful transapicalaortic valve implantation with the CoreValve Revalving system:a case report. Heart Surg Forum 2007;10:E478—9.

26] Petronio AS, De Carlo M, Bedogni F, et al. Safety and efficacy ofthe subclavian approach for transcatheter aortic valve implan-tation with the CoreValve revalving system. Circ CardiovascInterv 2010;3:359—66.

27] Modine T, Obadia JF, Choukroun E, et al. Transcutaneous aorticvalve implantation using the axillary/subclavian access: fea-sibility and early clinical outcomes. J Thorac Cardiovasc Surg2011;141:487—91.

28] Bapat V, Khawaja MZ, Attia R, et al. Transaortic tran-scatheter aortic valve implantation using Edwards Sapienvalve: a novel approach. Catheter Cardiovasc Interv 2011,doi:10.1002/ccd.23276. Epub ahead of print.

29] Etienne PY, Papadatos S, El Khoury E, Pieters D, Price J, GlineurD. Transaortic transcatheter aortic valve implantation withthe Edwards SAPIEN valve: feasibility, technical considerations,and clinical advantages. Ann Thorac Surg 2011;92:746—8.